摘要
研究一种基于全景视觉的机器人自主定位的方法。利用光学视觉原理,设计了一种全景视觉传感器,从而获得机器人周围环境的全方位景物的图像。通过去除全景图像的噪声、分割颜色阈值、计算中心点等处理,识别出机器人周围景物的已知路标,采用三角定位法,计算出机器人的坐标,从而为机器人的导航、避碰等任务奠定良好的基础。实验结果表明,此方法对于机器人自主定位具有一定的可行性。
A method based on the omni-vision was studied for the self-location of mobile robot. According to optical principle, the omni-vision sensor was designed. By using the omni-vision sensor, robot＇s panoramic image in working environment was obtained. With the aid of noise reduction, color threshold segmentation and center point computation and so on, the landmarks in the panoramic image were recognized. By using triangulation method, the robot＇s position was obtained, which was a good basis for the navigation and the collision of robot. Experimental results prove that the method is feasible for the self-location of mobile robot.
出处
《计算机应用》
CSCD
北大核心
2007年第7期1677-1679,共3页
journal of Computer Applications
关键词
全景视觉
移动机器人
路标识别
图像处理
omni-vision
mobile robot
landmark recognition
image processing
